Communication control device and communication control method

Abstract

To allocate radio resources to a plurality of communication terminals with a simple configuration.SOLUTION: A communication control device 1 for allocating radio resources to a plurality of communication terminals 2T each of which transmits an origination request signal at an arbitrary time includes: an acquisition part 11 for acquiring first information indicating the order of the origination request signals transmitted by each of the plurality of communication terminals 2T; a calculation part 13 for outputting second information regarding the order of origination signals transmission by the plurality of communication terminals 2T sharing the radio resources corresponding to origination request signals, and the order where the number and frequency of transmissions of the origination signal are uniform among the plurality of communication terminals 2T, by giving the first information as unknown input to a learned machine learning model and calculating the learned machine learning model; and a communication control part 15 for notifying each of the plurality of communication terminals 2T of the second information as radio resource allocation information.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a communication control device and a communication control method. [Background technology]

[0002] The spread of mobile devices such as smartphones and IoT devices that allow various things to communicate has created a need for communication control for a huge number of mobile devices and IoT devices. If a communication line were secured for each communication device such as a mobile device or IoT device, the number of lines would be huge, exceeding 1,000 lines, for example, and control and management of the lines would become complicated.

[0003] Therefore, multiple access using multiplexing technology has been used to share a single line among multiple communication terminals, and each communication terminal is controlled to connect independently. For example, in time division multiple access (TDMA) using time division multiplexing (TDM), a single frequency is divided into multiple parts, and the divided time, or time slots, are assigned to each communication terminal for communication. This improves the efficiency of communication lines, leading to cost reductions in equipment costs for base stations and core networks, as well as communication costs.

[0004] However, in conventional communication using time division multiple access, it was necessary to synchronize the time of the time slots between the communication terminal side and the base station and core network side with high accuracy. For example, Patent Document 1 discloses a technology that calculates the frequency deviation ratio between a TSN (Time-Sensitive Networking) clock and a clock that is the operation timing of a 5G system including a base station and a communication terminal, and corrects the time synchronization deviation between the two clocks based on the calculated frequency deviation.

[0005] However, in the technique disclosed in Patent Document 1, the system configuration becomes complicated in order to perform highly accurate synchronization between the communication terminal and the network side. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] JP 2023-058607 A Summary of the Invention [Problem to be solved by the invention]

[0007] In conventional techniques, it has not been possible to allocate radio resources to multiple communication terminals using a simple configuration without performing highly accurate synchronization processing.

[0008] The present invention has been made in order to solve the above-mentioned problems, and has an object to allocate radio resources to a plurality of communication terminals with a simple configuration. [Means for solving the problem]

[0009] In order to solve the above-mentioned problems, the communication control device of the present invention is a communication control device that allocates radio resources to a plurality of communication terminals, each of which transmits a call request signal at an arbitrary time, and includes a first acquisition unit configured to acquire first information indicating the order in which the call request signal was transmitted by each of the plurality of communication terminals, a calculation unit configured to provide the first information as an unknown input to a trained machine learning model, perform calculations on the trained machine learning model, and output second information regarding the order in which each of the plurality of communication terminals sharing the radio resources transmits a call signal corresponding to the call request signal, in which the number and frequency of transmitting the call signal are equal among the plurality of communication terminals, and a communication control unit configured to notify each of the plurality of communication terminals of the second information as allocation information of the radio resources.

[0010] In addition, the communication control device of the present invention further includes a confirmation unit configured to confirm in order for each communication terminal whether the call request signal has been transmitted by the multiple communication terminals; a second acquisition unit configured to acquire third information regarding an order in which the call request signal was transmitted by each of the multiple communication terminals, the order being based on the communication terminal for which it has been confirmed that the call request signal has been transmitted by the confirmation unit; a learning unit configured to learn, using a machine learning model, a relationship between the third information and an order in which each of the multiple communication terminals sharing the wireless resource transmits the call signal, the order in which the number and frequency of transmitting the call signal are equal among the multiple communication terminals other than the communication terminal when the communication terminal is used as a reference; and a memory unit configured to store the trained machine learning model constructed by the learning unit, and the calculation unit may read out the trained machine learning model from the memory unit and perform calculations on the trained machine learning model.

[0011] In the communication control device of the present invention, the third information may be an order based on each of the plurality of communication terminals.

[0012] In addition, in the communication control device of the present invention, the calculation unit may provide the third information as the unknown input to the trained machine learning model, perform calculations on the trained machine learning model, and output an order in which each of the multiple communication terminals sharing the wireless resources transmits the outgoing signal, and in which, when the communication terminal is used as a reference, the number and frequency of transmitting the outgoing signal are equal among the multiple communication terminals other than the communication terminal.

[0013] In addition, in the communication control device according to the present invention, the machine learning model may be a neural network having an input layer, a hidden layer, and an output layer.

[0014] In order to solve the above-mentioned problems, a communication control method according to the present invention is a communication control method for allocating radio resources to a plurality of communication terminals, each of which transmits a call request signal at an arbitrary time, and includes a first step of acquiring first information indicating the order in which the call request signal was transmitted by each of the plurality of communication terminals, a second step of providing the first information as an unknown input to a trained machine learning model, performing calculations on the trained machine learning model, and outputting second information regarding the order in which each of the plurality of communication terminals sharing radio resources transmits a call signal corresponding to the call request signal, in which the number and frequency of transmitting the call signal are equal among the plurality of communication terminals, and a third step of notifying each of the plurality of communication terminals of the second information as allocation information of the radio resources.

[0015] In addition, the communication control method of the present invention further includes a fourth step of checking in order for each communication terminal whether the call request signal has been transmitted by the multiple communication terminals; a fifth step of acquiring third information regarding the order in which the call request signal has been transmitted by each of the multiple communication terminals, the order being based on the communication terminal for which it has been confirmed in the fourth step that the call request signal has been transmitted; a sixth step of learning, using a machine learning model, a relationship between the third information and an order in which each of the multiple communication terminals sharing the wireless resource transmits the transmission signal, the order in which the number and frequency of transmitting the transmission signal are equal among the multiple communication terminals other than the communication terminal when the communication terminal is used as a reference; and a seventh step of storing the trained machine learning model constructed in the sixth step in a memory unit, and the second step may read out the trained machine learning model from the memory unit and perform calculations on the trained machine learning model.

[0016] In the communication control method according to the present invention, the third information may be an order based on each of the plurality of communication terminals.

[0017] Furthermore, in the communication control method of the present invention, the second step may include providing the third information as the unknown input to the trained machine learning model, performing calculations on the trained machine learning model, and outputting an order in which each of the multiple communication terminals sharing the wireless resources transmits the outgoing signal, in which, when the communication terminal is used as a reference, the number and frequency of transmitting the outgoing signal are equal among the multiple communication terminals other than the communication terminal. Effect of the Invention

[0018] According to the present invention, the trained machine learning model is calculated, and second information is output regarding an order in which each of a plurality of communication terminals sharing a wireless resource transmits an outgoing signal corresponding to an outgoing request signal, and the number and frequency of the outgoing signals are equal among the plurality of communication terminals. Therefore, wireless resources can be allocated to a plurality of communication terminals with a simpler configuration. [Brief description of the drawings]

[0019] [Figure 1] FIG. 1 is a block diagram showing a configuration of a communication control system including a communication control device according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a diagram for explaining an overview of the communication control device according to the present embodiment. [Diagram 3] FIG. 3 is a diagram for explaining the learning process performed by the learning unit according to the present embodiment. [Figure 4] FIG. 4 is a diagram for explaining the learning process performed by the learning unit according to the present embodiment. [Diagram 5] FIG. 5 is a block diagram showing a hardware configuration of the communication control device according to the present embodiment. [Figure 6] FIG. 6 is a flowchart showing the operation of the communication control device according to the present embodiment. [Figure 7] FIG. 7 is a flowchart showing the operation of the communication control device according to the present embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will now be described in detail with reference to FIGS.

[0021] [Communication control system configuration] First, an overview of a communication control system including a communication control device 1 according to an embodiment of the present invention will be described. As shown in FIG. 1, the communication control system includes a communication control device 1, a plurality of communication terminals 2T1, 2T2, . . . , 2T n (n=1, 2, . . .), a base station 3, and a core network 4. The communication control system includes, for example, a plurality of communication terminals 2T1, 2T2, . . . , 2T n When multiple communication terminals 2T1, 2T2, . . . , 2T n controls the order in which the outgoing signals are sent.

[0022] As shown in FIG. 1, communication terminals 2T1, 2T2, . . . , 2T n is realized by a mobile communication terminal such as a smartphone equipped with a SIM, a PDA (Personal Digital Assistant), a tablet computer, or a laptop computer (a so-called notebook computer). n This can also be achieved by IoT devices that have their own unique IP addresses and can connect to the Internet.

[0023] Communication terminals 2T1, 2T2,..., 2T n Each SIM is uniquely identified by an International Mobile Subscriber Identity (IMSI), which is a subscriber identification number that identifies the SIM.

[0024] Communication terminals 2T1, 2T2,..., 2T n When starting communication or in order to start transmission and reception of a transmission signal such as a data signal, the communication terminals 2T1, 2T2, . . . , 2Tn is a channel request signal for requesting a resource block of radio resources such as a time slot.

[0025] Communication terminals 2T1, 2T2,..., 2T n However, when the communication terminal 2T1 transmits a call request signal to the base station 3, the base station 3 assigns available wireless resources such as unused time slots to the communication terminals 2T1, 2T2, . . . , 2T n In this embodiment, the communication terminals 2T1, 2T2, . . . , 2T n The order in which the terminals transmit signals such as data signals is notified as information corresponding to the allocation of time slots.

[0026] For example, communication terminals 2T1, 2T2, . . . , 2T n The communication terminals 2T1, 2T2, . . . , 2T3 transmit control signals processed in a C-Plane (Control Plane) and data signals processed in a U-Plane (User Plane) to a core network 4 via a base station 3. n The data includes user data such as text, voice, and images transmitted by the communication terminals 2T1, 2T2, . . . , 2T n When no distinction is made between them, they may be collectively referred to as the communications terminal 2T.

[0027] The base station 3 is composed of a wireless base station compatible with LTE wireless communication and a wireless base station compatible with 5G. The base station 3 is connected to communication terminals 2T1, 2T2, . . . , 2T n and the core network 4. In this embodiment, the communication control device 1 and the base station 3 are connected to each other so that they can communicate with each other via the core network 4 and the network (NW). Furthermore, the base station 3 relays communication between the communication terminals 2T1, 2T2, . . . , 2Tn This holds the transmission timestamp (in ms) of the transmission request signal sent by the

[0028] The core network 4 is composed of a 5G SA (Stand Alone) core network and a NSA (Non Stand Alone) core network. n It has the functions of processing outgoing signals such as control signals and data signals transmitted via base station 3, managing communication protocols and security functions, and ensuring communication quality and safety, such as communication speed.

[0029] Fig. 2(a) is a schematic diagram showing allocation of radio resources to a plurality of communication terminals 2T by a communication control system according to a conventional example. Fig. 2(b) is a schematic diagram showing a configuration in which a communication control system including a communication control device 1 according to the present embodiment allocates radio resources to a plurality of communication terminals 2T.

[0030] In (a) and (b) of Fig. 2, the "terminal side" shown on the left side is the allocation of time slots as seen from the communication terminal 2T side. Meanwhile, the "network side" shown on the right side indicates the allocation of time slots to the communication terminal 2T as seen from the base station 3 and the core network 4, i.e., the communication control system according to this embodiment. Also, the numbers "1, 2, 3,..., l,..., m,...n" shown in (a) and (b) of Fig. 2 indicate the allocation of time slots to n communication terminals 2T1, 2T2,..., 2T n Indicates the index number of the

[0031] In a conventional communication control system shown in Fig. 2(a), n time slots corresponding to the time obtained by dividing one frequency into n parts by the TDMA method are fixedly assigned to each communication terminal 2T. In Fig. 2(a), the timing of signal transmission and reception between the communication terminal 2T and the base station 3 is assigned to each time slot, and the timing of transmission and reception of signals between the communication terminal 2T and the base station 3 is controlled so as to match.

[0032] On the other hand, the communication control device 1 according to the present embodiment shown in FIG. 2(b) has n communication terminals 2T1, 2T2, . . . , 2T n When sharing one frequency, one line is divided into n time slots, and n communication terminals 2T1, 2T2,...,2T n In addition, when the communication terminal 2T (for example, the communication terminal 2T1) that transmitted the transmission request signal satisfies a predetermined condition, the time slot with the smallest number is assigned to the communication terminal 2T1, and the communication terminal 2T1 can transmit the transmission signal first.

[0033] In addition, in the communication control system according to the present embodiment, a communication control device 1 provided on the network side uses a learned machine learning model to determine whether a call request signal is transmitted from a plurality of communication terminals 2T1, 2T2, . . . , 2T n determines the order in which to transmit signals such as control signals and data signals. n The order in which the transmission signals are transmitted corresponds to the time slot numbers of the n consecutive time slots.

[0034] In the communication control system according to the present embodiment, n communication terminals 2T1, 2T2, . . . , 2T n For example, the controller 101 cycles through the communication terminals 2T1, one by one in the order of index values ​​1 to n. For example, while the controller 101 cycles through the communication terminals 2T1, it checks whether the communication terminal 2T1 has transmitted a call request signal, and similarly, while the controller 101 cycles through the communication terminals 2T2, it checks whether the communication terminal 2T2 has transmitted a call request signal.

[0035] Here, the communication terminals 2T1 to 2T n In the first round, the communication terminals 2T1, 2T2, . . . , 2T n The time slot with the smallest time slot number is assigned to the communication terminal 2T that has been confirmed to have transmitted a transmission request signal at the earliest time. In this case, the n consecutive time slots are assigned to the n communication terminals 2T1, 2T2, . . . , 2T nThere are n! possible combinations of orders for assigning all of .

[0036] However, controlling transmission using all n! combinations of the order in which time slots are allocated is a burden on the base station 3 and the core network 4. Also, for example, even if it is detected that a communication terminal 2T1 has transmitted a transmission request signal during a period in which the communication terminal 2T1 is visiting another communication terminal 2T1, there may be cases in which another communication terminal 2T2 has also transmitted a transmission request signal during the same period. In such cases, it is necessary to adjust the transmission order between the communication terminals 2T1 and 2T2.

[0037] In the communication control system according to the present embodiment, when a predetermined communication terminal 2T in a patrol transmits a transmission request signal, the time slot with the smallest time slot number is basically assigned, and n communication terminals 2T1, 2T2, . . . , 2T n In order to fairly distribute the load among communication terminals 2T1, 2T2, . . . , 2T n determines the order in which the outgoing signals are transmitted, i.e., the order in which time slots are assigned.

[0038] More specifically, in the communication control system according to the present embodiment, a communication control device 1 provided on the network side uses a trained machine learning model to control a plurality of communication terminals 2T1, 2T2, . . . , 2T n determines the order in which to transmit the outgoing signals.

[0039] The trained machine learning model is a set of communication terminals 2T1, 2T2, . . . , 2T n The order in which the communication terminals 2T1, 2T2, . . . , 2T n This model learns in advance the relationship between the number of times and the frequency of transmitting signals such as data signals between communication terminals 2T1, 2T2, . . . , 2T nFor example, by using only n predetermined combinations, the processing load of the base station 3 and the core network 4 on the transmission control can be reduced.

[0040] In the communication control system according to the present embodiment, communication terminals 2T1, 2T2, . . . , 2T n The time slots are assigned according to the order of transmitting the transmission signal determined based on the confirmation by the communication terminal 2T cyclically checking whether the communication terminal 2T has transmitted a transmission request signal. Therefore, there is no need for highly accurate synchronization between the communication terminal 2T and the base station 3, which was required in the conventional example.

[0041] [Function block of communication control device] The communication control device 1 includes a confirmation unit 10, an acquisition unit 11, a learning unit 12, a calculation unit 13, a machine learning (ML) model storage unit 14, a communication control unit 15, and a presentation unit 16. The communication control device 1 performs calculations of a trained machine learning model, and provides a plurality of communication terminals 2T1, 2T2, . . . , 2T n determines the order in which each of them emits an emission signal.

[0042] The confirmation unit 10 is a unit for confirming whether a plurality of communication terminals 2T1, 2T2, . . . , 2T n A transmission request signal transmitted at an arbitrary time from each of the communication terminals 2T1, 2T2, . . . , 2T n Whether the signal was sent by the communication terminals 2T1, 2T2, . . . , 2T n More specifically, as shown by the circular arrow in FIG. 2B, the confirmation unit 10 checks, for example, n communication terminals 2T1, 2T2, . . . , 2T n The system goes through each one and checks whether a call request signal has been sent.

[0043] The acquisition unit 11 (first acquisition unit) acquires a plurality of communication terminals 2T1, 2T2, . . . , 2T nThe acquiring unit 11 acquires information (first information) indicating the order in which the call request signals were transmitted by each of the acquiring unit 11 and the first information acquired by the acquiring unit 11. The first information acquired by the acquiring unit 11 is used as an unknown input when the later-described computing unit 13 computes a trained machine learning model.

[0044] In addition, the acquisition unit 11 (second acquisition unit) acquires the order based on the communication terminal 2T that has been confirmed by the confirmation unit 10 to have transmitted the transmission request signal, and n and obtains information (third information) regarding the order in which the call request signals were sent by each of the above.

[0045] The third information acquired by the acquisition unit 11 is used as data constituting teacher data when performing a learning process of a machine learning model in a learning unit 12 described later. Note that a calculation unit 13 described later can use not only the first information but also the third information as unknown input when calculating a trained machine learning model.

[0046] Regarding the third information, for example, a plurality of communication terminals 2T1, 2T2, . . . , 2T n During the period in which confirmation unit 10 is cycling through communication terminal 2T1, it is confirmed that communication terminal 2T1 has transmitted a transmission request signal. In this case, communication terminal 2T1 is determined to be the first of the first to nth orders for transmitting a transmission signal. Acquisition unit 11 acquires combinations of orders from second to nth, with communication terminal 2T1 being the first. Therefore, when communication terminal 2T1 is taken as the representative of n terminals, there are (n-1)! possible combinations of orders.

[0047] In this way, the order based on the communication terminal 2T means a combination of the order of the first to nth communication terminals 2T in the order when the time slot with the smallest number is assigned to the communication terminal 2T that is confirmed to have transmitted a transmission request signal in the round of the confirmation unit 10. Furthermore, the acquisition unit 11 acquires n communication terminals 2T1, 2T2, . . . , 2T n Obtain an order combination based on each of the above once.

[0048] The information on the order in which the call request signals were transmitted, which is acquired by the acquisition unit 11, is, for example, 1st: communication terminal 2T1, 2nd: communication terminal 2T2, . . . , nth: communication terminal 2T n The information is such that the identification number of each communication terminal 2T is associated with a value indicating the order or a time slot number.

[0049] The acquisition unit 11 can acquire the order of transmission based on the transmission timestamp (in ms) which is the date and time of transmission of the transmission request signal. For example, the acquisition unit 11 can acquire the transmission timestamp of the transmission request signal held in the base station 3 via the network NW.

[0050] The learning unit 12 receives the third information and the communication terminals 2T1, 2T2, . . . , 2T n is an order in which each of the communication terminals 2T1, . . . , 2T2, . . . , 2T3 transmits a transmission signal such as a data signal related to a transmission request signal, and the number and frequency of transmitting a transmission signal are determined based on the communication terminal 2T1, which has been confirmed to have transmitted a transmission request signal by the confirmation unit 10. n The machine learning model is used to learn the relationship between the order and the equal order between the two.

[0051] The learning unit 12 is configured to learn the communication terminals 2T1, 2T2, . . . , 2T n The number and frequency of transmitting a transmission signal such as a data signal are determined for each combination of the order in which the communication terminals 2T1, 2T2, . . . , 2T n The machine learning model is trained to produce an evenly balanced combination of orders.

[0052] The number and frequency of transmitting signals are equal for communication terminals 2T1, 2T2, . . . , 2T n The order is n communication terminals 2T1, 2T2, . . . , 2T nEach of these refers to a combination of the order in which the number and frequency of transmissions is equal for the first, second, ..., nth. As mentioned above, the order in which the transmission signals are transmitted corresponds to the time slot number.

[0053] For example, the first communication terminal 2T1 transmits a transmission signal at the earliest time using the first time slot. The second communication terminal 2T2 transmits a transmission signal using the second time slot after the communication terminal 2T1 transmits a transmission signal. Similarly, the third to n-th communication terminals 2T3 to 2T n Regarding the above, following the transmission of a transmission signal from the immediately preceding communication terminal 2T, the communication terminals 2A and 2B can sequentially transmit a transmission signal.

[0054] FIG. 3 is a diagram for explaining the teacher data used by the learning unit 12 in the learning process. The left column of FIG. 3 shows input values ​​that the learning unit 12 gives to the machine learning model. The right column of FIG. 3 shows the correct answer labels of the teacher data used by the learning unit 12 in the learning process. In the example of FIG. 3, communication terminals 2T1, 2T2, . . . , 2T n 3 shows the permutation of the order in which the four communication terminals 2T1, 2T2, 2T3, and 2T4 communicate when there are four communication terminals (n=4). The numbers "1", "2", "3", and "4" in Fig. 3 show the index values ​​of the four communication terminals 2T1, 2T2, 2T3, and 2T4.

[0055] As shown in the left column of FIG. 3, n communication terminals 2T1, 2T2, . . . , 2T n The permutation P of the n communication terminals 2T1, 2T2, . . . , 2T n When each of the communication terminals 2T1 and 2T2 is first to have the opportunity to transmit, the combination of the order is expressed as {(n-1)×(n-2)×1}. When n=4, if the communication terminal 2T1 is defined as the first, there are six combinations of the order.

[0056] In this embodiment, all n communication terminals 2T1, 2T2, . . . , 2Tn The combination of orders in which the communication terminal 2T is always designated as the first one at least once is used as the correct label shown in the right column of Fig. 3. Furthermore, in the second and subsequent orders, each communication terminal 2T other than the communication terminal 2T designated as the first one is designated as the second to nth one once each. In other words, from the combination of orders of {(n-1) × (n-2) × 1}, n communication terminals 2T1, 2T2, ..., 2T n The representative combination selected so that each has an equal opportunity to transmit a signal is regarded as the correct label.

[0057] 4 shows a neural network structure that is adopted as an example of a machine learning model for learning by the learning unit 12. The neural network includes an input layer x, a hidden layer h, and an output layer y. In the example of FIG. 4, the number of input nodes and output nodes are the same, and these are connected to communication terminals 2T1, 2T2, . . . , 2T n The number n of communication terminals 2T adopted as the number of input nodes and output nodes can be set according to the facility capacity of the base station 3, the core network 4, and the like.

[0058] The learning unit 12 detects whether a transmission request signal that each of the communication terminals 2T1, 2T2, . . . , 2T n The order of the signals sent by each of the input nodes x1 to x2 is fed to the input layer of the neural network, an activation function is applied to the weighted sum of the inputs, and the output determined by threshold processing is sent to the output layer. Using the example of Figure 3, n The values ​​of the identification information of the communication terminal 2T that transmitted the first transmission request signal to the n-th transmission request signal, which are indicated by "input value", are input to each of the output nodes y1 to y n is the output value of the neural network for the “input value”.

[0059] The size and elements of the machine learning model, including the number of layers in the hidden layer h and the density of the connections between nodes in the neural network, are not limited as long as the design provides sufficient inference accuracy. For example, the connections between nodes may be fully connected or sparsely connected.

[0060] The learning unit 12 further introduces an objective function that evaluates the error between the value of the correct label and the output value y of the neural network, thereby learning n communication terminals 2T1, 2T2, . . . , 2T n Each of the communication terminals 2T1, 2T2, . . . , 2T n The neural network parameters are trained so that the order of the labels corresponds to the order of the correct labels. To explain this using the example of Fig. 3, the training unit 12 trains the neural network using training data in which the "correct labels" in the left column are assigned to the "input values" enclosed in the dashed squares.

[0061] In this way, the “input value” in the teacher data is the number of communication terminals 2T2, . . . , 2T when the communication terminal 2T that has been confirmed by the confirmation unit 10 to have transmitted a transmission request signal is used as a reference. n In other words, in the example of FIG. 3, when the communication terminal 2T1 is defined as the first communication terminal, the information (third information) on the order in which the call request signal was transmitted by each of the communication terminals 2T2, . . . , 2T n The ordered combination of can be treated as a single group of input values.

[0062] In addition, the “correct label” is, for example, the nth communication terminal 2T2, . . . , 2T n In the combination of the order of communication terminals 2T2, . . . , 2T n Therefore, in the example of FIG. 3, when the communication terminal 2T1 is defined as the first, the first n communication terminals 2T2, . . . , 2T n The correct label will be assigned to a group of input values ​​for each combination of the above orders.

[0063] The learning unit 12 adjusts the weight parameters of the neural network so that the objective function is minimized, that is, becomes 0. The learning unit 12 can optimize the objective function by a gradient method using backpropagation, stochastic gradient descent, or the like. The trained machine learning model constructed by the learning unit 12 is stored in the ML model storage unit 14, which will be described later.

[0064] Returning to FIG. 1, the calculation unit 13 calculates the number of communication terminals 2T1, 2T2, . . . , 2T n Information (first information) regarding the order in which the transmission request signals were transmitted by each of the above is provided to the trained machine learning model as an unknown input.

[0065] The calculation unit 13 performs calculations on the trained machine learning model to calculate the wireless resource sharing among the communication terminals 2T1, 2T2, . . . , 2T n , 2T n The second information output section 11 outputs information (second information) regarding the order in which the number and frequency of transmitting the transmission signal are equalized between the first and second terminals.

[0066] The information on the order in which the call request signals were sent, which is given as an unknown input by the calculation unit 13, is n The order in which each of the communication terminals 2T1, 2T2, . . . , 2T n For example, there may be a case where time slots are allocated disproportionately to a particular communication terminal 2T1.

[0067] The calculation unit 13 calculates the number and frequency of transmitting the transmission signal by the calculation of the trained machine learning model. n It is possible to output a combination of call orders that are equal among the

[0068] In addition to the case where the calculation unit 13 provides the first information as an unknown input to the trained machine learning model, the calculation unit 13 can be configured to provide the third information as an unknown input to the trained machine learning model. In this case, the calculation unit 13 can use, as the unknown input, the order of transmission of the first to n-th transmission request signals, starting with the communication terminal 2T that has been confirmed to have transmitted a transmission request signal by the confirmation unit 10 making a round.

[0069] The ML model storage unit 14 stores the learned parameters constructed by the learning unit 12.

[0070] The communication control unit 15 controls the communication terminals 2T1, 2T2, . . . , 2T n , 2T1, 2T2, . . . , 2T3, 2T4, . . . n The time slot number based on the order in which the transmission signal is to be transmitted can be notified to each of the other stations.

[0071] The communication terminals 2T1, 2T2, . . . , 2T n Specifically, the communication terminals 2T1, 2T2, . . . , 2T n can each transmit a signal in a corresponding time slot in a specified order.

[0072] The presentation unit 16 is configured to receive the information from the communication terminals 2T1, 2T2, . . . , 2T n The presentation unit 16 presents information on the order of transmitting the transmission signals, which is notified to the user. The presentation unit 16 can also present information on communication control, such as allocation of time slots.

[0073] [Hardware configuration of communication control device] Next, an example of a hardware configuration for realizing the communication control device 1 having the above-mentioned functions will be described with reference to FIG.

[0074] As shown in FIG. 5, the communication control device 1 can be realized, for example, by a computer having a processor 102, a main memory device 103, a communication interface 104, an auxiliary memory device 105, and an input / output (I / O) 106 connected via a bus 101, and a program that controls these hardware resources.

[0075] The main memory device 103 prestores programs for the processor 102 to perform various controls and calculations. The processor 102 and the main memory device 103 realize the functions of the communication control device 1, such as the confirmation unit 10, the acquisition unit 11, the learning unit 12, the calculation unit 13, and the communication control unit 15 shown in FIG.

[0076] The communication interface 104 is an interface circuit for network connection between the communication control device 1 and various external electronic devices.

[0077] The auxiliary storage device 105 is composed of a readable / writable storage medium and a drive for reading and writing various information such as programs and data from and to the storage medium. The auxiliary storage device 105 can use a semiconductor memory such as a hard disk or a flash memory as the storage medium.

[0078] The auxiliary storage device 105 has a program storage area for storing a communication control program executed by the communication control device 1. The auxiliary storage device 105 also has an area for storing a learning program for learning a machine learning model. The auxiliary storage device 105 realizes the ML model storage unit 14 described in Fig. 1. Furthermore, the auxiliary storage device 105 may have, for example, a backup area for backing up the above-mentioned data, programs, and the like.

[0079] The input / output I / O 106 is an input / output device that inputs signals from an external device and outputs signals to an external device.

[0080] [Operation of communication control device] Next, the operation of the communication control device 1 having the above-mentioned configuration will be described with reference to the flowcharts of Fig. 6 and Fig. 7. Fig. 6 is a flowchart showing the learning process by the communication control device 1. Fig. 7 shows the operation of the communication control device 1 by the computation of a trained machine learning model, which is performed on a plurality of communication terminals 2T1, 2T2, ..., 2T n 10 is a flowchart showing a communication control process for allocating wireless resources to a

[0081] First, as shown in Fig. 6, the learning unit 12 prepares teacher data (step S1). More specifically, first, the confirmation unit 10 checks a plurality of communication terminals 2T1, 2T2,..., 2T n , 2T1, 2T2, . . . , 2T3, 2T4, 2T5, 2T6, 2T7, 2T8, 2T9, 2T10, 2T11, 2T12, . . . , 2T13, 2T14, 2T15, 2T16, 2T17, 2T18, 2T19, 2T2 n The learning unit 12 uses the third information on the order acquired by the acquisition unit 11 as an input value constituting the teacher data. As described above, there are n pieces of third information.

[0082] Furthermore, the learning unit 12 learns a plurality of communication terminals 2T1, 2T2, . . . , 2T n is an order in which each of the communication terminals 2T1, . . . , 2T2, . . . is a sequence in which the communication terminal 2T1 confirmed by the confirmation unit 10 is used as a reference, and the number and frequency of transmitting the transmission signal are determined by the number of communication terminals 2T2, . . . , 2T3, . n The order that is equal among the input values ​​is assigned as the correct label.

[0083] The learning unit 12 learns all n communication terminals 2T1, 2T2, . . . , 2T n is always designated as the first one at least once, and in the second and subsequent orders, each of the communication terminals 2T other than the communication terminal 2T designated as the first one uses the combination of the second to nth designated orders once each as the correct label.

[0084] Furthermore, the learning unit 12 employs a neural network as a machine learning model, and sets the settings of the input layer, hidden layer, and output layer, as well as the initial values ​​of weight parameters, thresholds, and other parameters.

[0085] Next, the learning unit 12 performs a learning process using the teacher data prepared in step S1 (step S2). n and the order in which the call request signals are transmitted by the communication terminals 2T1, 2T2, . . . , 2T n is an order in which each of the communication terminals 2T1, 2T2, . . . , 2T n The system uses a neural network to learn the relationship between the order and the equality between the players.

[0086] More specifically, the learning unit 12 performs a process of learning the information (third information) acquired by the acquisition unit 11 and the communication terminals 2T1, 2T2, . . . , 2T n is an order in which each of the communication terminals 2T2, . . . , 2T1 transmits a transmission signal such as a data signal, and when a communication terminal 2T (for example, a communication terminal 2T1) that has been confirmed by the confirmation unit 10 to have transmitted a transmission request signal is used as a reference, the number and frequency of transmitting a transmission signal are determined based on the number and frequency of the communication terminals 2T2, . . . , 2T1 other than the communication terminal 2T1. n The machine learning model is used to learn the relationship between the order and the equal order between the two.

[0087] The learning unit 12 employs a machine learning model having a neural network structure as shown in FIG. n The weight parameters that minimize the objective function are repeatedly adjusted and updated so that the values ​​of the identification information of the communication terminals 2T that transmitted a call request signal first through the nth call request signal, which are input to the above, become the value of the correct label, and the values ​​of the weight parameters are determined.

[0088] Thereafter, the ML model storage unit 14 stores the learned weight parameters obtained by the learning process in step S2 (step S3).

[0089] Next, a control process using the trained neural network will be described with reference to FIG. 7. First, the acquisition unit 11 acquires an unknown input value to be given to the trained neural network (step S10). More specifically, the acquisition unit 11 acquires an unknown input value to be given to the trained neural network from the communication terminals 2T1, 2T2, . . . , 2T n The information regarding the order in which the transmission request signals were transmitted by each of the nodes is obtained as an unknown input value.

[0090] Alternatively, the acquisition unit 11 may be configured to acquire, as an unknown input value, the order of transmissions from the communication terminal 2T whose transmission of a transmission request signal has been confirmed by the rounds of the confirmation unit 10 up to the nth transmission.

[0091] Next, the calculation unit 13 loads the learned parameters from the ML model storage unit 14 (step S11). After that, the calculation unit 13 calculates the unknown input value acquired in step S10 as follows: The result is given to the trained neural network, and the trained neural network is operated (step S12).

[0092] The calculation unit 13 calculates the number of communication terminals 2T1, 2T2, . . . , 2T n is an order in which each of the communication terminals 2T1, 2T2, . . . , 2T n Outputs information about the order that is evenly matched between the

[0093] Next, the communication control unit 15 receives the communication terminals 2T1, 2T2, . . . , 2T n The order in which each of the communication terminals 2T1, 2T2, . . . , 2T n (step S13).

[0094] Thereafter, the presentation unit 16 presents the information relating to the communication control in step S13 to an external server or the like (step S14).

[0095] In the above embodiment, the case of a communication control system conforming to LTE and 5G is exemplified. However, the predetermined communication standard is not limited to LTE and 5G, and may be a communication control system conforming to 6G.

[0096] In the above embodiment, the learning unit 12 performs the learning process using a neural network as a machine learning model. However, the machine learning model may be a neural network model as described above, or a logistic regression, a random forest, a decision tree, or a deep learning model in which a neural network is multi-layered. In addition to these supervised learning models, a generative model such as a generative adversarial network or a variational autoencoder may be used as a machine learning model for performing unsupervised learning.

[0097] In the embodiment described above, the learning unit 12 performs the learning process, and the communication terminals 2T1, 2T2, . . . , 2T n In the above description, the calculation unit 13 that calculates the order in which the devices transmit outgoing signals, and the communication control unit 15 are all mounted on the communication control device 1. However, the learning unit 12, the calculation unit 13, and the communication control unit 15 may be provided as the same hardware configuration, or the learning process and the communication control process may be distributed to different servers on the network NW.

[0098] As described above, the communication control device 1 according to the present embodiment performs calculations of a trained machine learning model to control a plurality of communication terminals 2T1, 2T2, . . . , 2T n , 2T nTherefore, it is possible to allocate radio resources to a plurality of communication terminals with a simpler configuration.

[0099] The above describes the embodiments of the communication control device and communication control method of the present invention. However, the present invention is not limited to the described embodiments, and various modifications that may be envisioned by a person skilled in the art can be made within the scope of the invention described in the claims. [Explanation of symbols]

[0100] 1... communication control device, 10... confirmation unit, 11... acquisition unit, 12... learning unit, 13... calculation unit, 14... ML model storage unit, 15... communication control unit, 16... presentation unit, 2T, 2T1, 2T2,..., 2T n ...communication terminal, 3...base station, 4...core network, 101...bus, 102...processor, 103...main memory device, 104...communication interface, 105...auxiliary memory device, 106...input / output I / O, NW...network.

Claims

1. A communication control device that allocates radio resources to a plurality of communication terminals each of which transmits a call request signal at an arbitrary time, comprising: a first acquisition unit configured to acquire first information indicating an order in which the call request signals were transmitted by each of the plurality of communication terminals; a calculation unit configured to provide the first information as an unknown input to a trained machine learning model, perform calculations on the trained machine learning model, and output second information regarding an order in which each of the multiple communication terminals sharing a wireless resource transmits an outgoing signal corresponding to the outgoing request signal, the order being such that the number and frequency of transmitting the outgoing signal are equal among the multiple communication terminals; a communication control unit configured to notify each of the plurality of communication terminals of the second information as allocation information of the radio resources; A communication control device comprising:

2. 2. The communication control device according to claim 1, a confirmation unit configured to confirm, in order for each of the plurality of communication terminals, whether or not the call request signal has been sent by the communication terminal; a second acquisition unit configured to acquire third information regarding an order in which the call request signal was transmitted by each of the plurality of communication terminals, the order being based on the communication terminal in which it was confirmed by the confirmation unit that the call request signal was transmitted; a learning unit configured to learn, using a machine learning model, a relationship between the third information and an order in which each of the plurality of communication terminals sharing the wireless resource transmits the transmission signal, the order being an order in which the number and frequency of transmitting the transmission signal is equal among the plurality of communication terminals other than the communication terminal when the communication terminal is used as a reference; A storage unit configured to store the trained machine learning model constructed by the learning unit; Equipped with The calculation unit reads out the trained machine learning model from the storage unit and performs calculations on the trained machine learning model. A communication control device comprising:

3. 3. The communication control device according to claim 2, The third information is an order based on which each of the plurality of communication terminals is arranged. A communication control device comprising:

4. 3. The communication control device according to claim 2, The calculation unit provides the third information as the unknown input to the trained machine learning model, performs calculations on the trained machine learning model, and outputs an order in which each of the plurality of communication terminals sharing the wireless resource transmits the outgoing signal, and in which the number and frequency of transmitting the outgoing signal are equal among the plurality of communication terminals other than the communication terminal when the communication terminal is used as a reference. A communication control device comprising:

5. 5. The communication control device according to claim 4, The machine learning model is a neural network having an input layer, a hidden layer, and an output layer. A communication control device comprising:

6. A communication control method for allocating radio resources to a plurality of communication terminals each of which transmits a call request signal at an arbitrary time, comprising the steps of: a first step of acquiring first information indicating an order in which the call request signals were transmitted by each of the plurality of communication terminals; a second step of providing the first information as an unknown input to a trained machine learning model, performing a calculation on the trained machine learning model, and outputting second information regarding an order in which each of the plurality of communication terminals sharing a wireless resource transmits an outgoing signal corresponding to the outgoing request signal, the order being such that the number and frequency of transmitting the outgoing signal are equal among the plurality of communication terminals; a third step of notifying each of the plurality of communication terminals of the second information as allocation information of the radio resources; A communication control method comprising:

7. 7. The communication control method according to claim 6, a fourth step of checking in turn for each of the plurality of communication terminals whether or not the call request signal has been sent by the communication terminal; a fifth step of acquiring third information on an order in which the call request signal was transmitted by each of the plurality of communication terminals, the order being based on the communication terminal in which it was confirmed in the fourth step that the call request signal was transmitted; a sixth step of learning, using a machine learning model, a relationship between the third information and an order in which each of the plurality of communication terminals sharing the wireless resource transmits the transmission signal, the order being an order in which the number and frequency of transmitting the transmission signal is equal among the plurality of communication terminals other than the communication terminal when the communication terminal is used as a reference; A seventh step of storing the trained machine learning model constructed in the sixth step in a storage unit; Equipped with The second step reads out the trained machine learning model from the storage unit and performs a calculation on the trained machine learning model. A communication control method comprising:

8. 8. The communication control method according to claim 7, The third information is an order based on which each of the plurality of communication terminals is arranged. A communication control method comprising:

9. 8. The communication control method according to claim 7, The second step provides the third information as the unknown input to the trained machine learning model, performs a calculation on the trained machine learning model, and outputs an order in which each of the plurality of communication terminals sharing the wireless resource transmits the outgoing signal, and in which the number and frequency of transmitting the outgoing signal is equal among the plurality of communication terminals other than the communication terminal when the communication terminal is used as a reference. A communication control method comprising:

Images